Histocompatibility antigens are a group of cell membrane alloantigens that are recognized by T lymphocytes and thereby initiate graft rejection or graft-versus-host disease (GVHD) following transplantation (1). In the early days of immunogenetics, the identification of major histocompatibility complex (MHC) antigens was based on their strong immunogenicity in skin transplant experiments between congenic-resistant strains of mice. Other less potent antigens were called minor histocompatibility antigens (MiHA). It soon became obvious that a distinction between major and minor antigens based solely on their immunogenicity was imprecise, as some MHC antigens are weak immunogens while some MiHA appear “neither weak nor minor” (2;3) It is now known that MHC antigens (also referred to as HLA antigens) are transmembrane glycoproteins encoded by closely linked polymorphic loci located on chromosome 6 in humans. Their primary role is to bind endogenous and exogenous peptides that are scrutinized by T cells. MHC (or HLA) molecules present thousands of peptides at the surface of human cells (4;5). These MHC-associated peptides (MAPs) are referred to as the immunopeptidome and they originate from proteasomal processing and further processing of endogenous proteins (6-8). The immunopeptidome of identical twins (also referred to as syngeneic subjects) is identical. By contrast, MAPs present on cells from HLA-identical non-syngeneic subjects are classified into two categories: i) invariant MAPs which are present in all subjects with a given HLA type, and ii) MiHAs which are MAPs present in some but absent in other subjects (9). When T cells are transplanted into an MHC-identical host, they react promptly and specifically to what they see as non-self: host-specific MiHAs. MiHAs are essentially genetic polymorphisms that are immunogenic for T cells. MiHAs are a consequence of any form of accumulated genetic variation that translates to differential MAP display (3; 9-13).
Two main strategies can be used for cancer immunotherapy: vaccination and adoptive T-cell immunotherapy (ATCI). The term “ATCI” refers to transfusion of T lymphocytes that may come from different types of donors: the patient (autologous), a genetically-identical twin (syngeneic), or a non-identical donor (allogeneic). To date, ATCI has yielded much higher cancer remission and cure rates than vaccines, and the most widely used form of cancer ATCI is allogeneic hematopoietic cell transplantation (AHCT) (17-22). The graft-versus-tumor (GVT) effect induced by allogeneic AHCT is due mainly to T-cell responses against host MiHAs: GVT is abrogated if the donor is an identical twin (no MiHA differences with the recipient) or if the graft is T-cell depleted (20;23). More than 200,000 individuals treated for hematological malignancies owe their life to the MiHA-dependent GVT effect which represents the most striking evidence of the ability of the human immune system to eradicate neoplasias (1824-28). Though the allogeneic GVT effect is being used essentially to treat patients with hematologic malignancies, preliminary evidence suggests that it may be also effective for the treatment of solid tumors (29-33). Nonetheless, the considerable potential of MiHA-targeted cancer immunotherapy has not been properly exploited in medicine. In current medical practice, MiHA-based ATCI is limited to “conventional” AHCT, that is, injection of hematopoietic cells from an allogeneic HLA-matched donor. Such unselective injection of allogeneic lymphocytes is a very rudimentary form of MiHA-targeted therapy. First, it lacks specificity and is therefore highly toxic: unselected allogeneic T cells react against a multitude of host MiHAs and thereby induce GVHD in 60% of recipients. GVHD is always incapacitating and frequently lethal (34-38). Second, conventional AHCT induces only an attenuated form of GVT reaction because donor T cells are not being primed (pre-activated) against specific MiHAs expressed on cancer cells prior to injection into the patient. While primed T cells are resistant to tolerance induction, naïve T cells can be tolerized by tumor cells (39-42).
It has been demonstrated in mice models of AHCT that, by replacing unselected donor lymphocytes with CD8 T cells primed against a single MiHA, it was possible to cure leukemia and solid tumors without causing GVHD or any untoward effect (33; 43; 44). Success depends on two main points: selection of an immunodominant (highly immunogenic) MiHA expressed on neoplastic cells, and priming of donor CD8 T cells against the target MiHA prior to AHCT. A recent article (20) describes why MiHA-targeted ATCI is so effective and how translation of this approach in the clinic could have a significant impact on cancer immunotherapy. Implementation of MiHA-targeted ATCI in humans has been limited mainly by the paucity of molecularly defined human MiHAs. Thus, only 33% of patients with leukemia would be eligible for MiHA-based ATCI (15).
Human MiHAs have been discovered using reductionist T-cell based methods. Starting with cytotoxic T lymphocytes (CTLs) from an individual reactive against cells of another HLA-identical subject, investigators have tried and identified MiHAs recognized by these T cells. Different methods have used to do so. First, CTLs were tested on MiHA-negative cells coated with MAPs eluted from MiHA-positive cells. The MAP eluates were fractionated and ultimately the MiHA recognized by CTLs was sequenced by mass spectrometry (MS) (48-53). Second, CTLs were used to screen MiHA-negative cells transfected with cDNA libraries to identify MiHA-coding transcripts (16; 54-59). Finally, CTLs have been tested on lymphoblastoid cell lines from many subjects and linkage analyses were performed (based for instance on whole genome association scans or HapMap resources) on lines recognized or not by CTLs (60-67).
The various methods used to discover MiHAs present significant caveats. Firstly, they are not really suitable for high-throughput MiHA discovery: MiHA discovery is made one by one and depends on the availability of a CTL line. Secondly, only MiHAs that have been eluted from living cells and identified by MS can be considered to be validated (direct identification). In the other cases (indirect identification), uncertainty remains as to the exact structure of MiHAs naturally presented at the cell surface (an important criterion for MiHA-targeted immunotherapy). The ambiguity stems mainly from two factors: i) T cells are eminently cross-reactive and can recognize more than one peptide (68); ii) bioinformatic tools used for identification of MAPs in general and MiHAs in particular do not have sufficient reliability to replace direct proteomic identification (69-71).
Thus, there is a need for novel approaches for the identification of MiHAs.
The present description refers to a number of documents, the content of which is herein incorporated by reference in their entirety.